Jean-Marie Zajac pointed out that accented characters in ISO-8859-1
encoding are not displayed as expected. Scamper is not encoding-aware,
but it translates ISO-8859-1 to the encoding that is used in Squeak.
Unfortunately, due to a subtle bug the translation is done twice:
First, the entire source string is translated, later parsed entities are
translated again.

This change set drops the translation of parsed entites. To make things
work,
it adds the translation of character entity references (characters that are
form &#<integer>; or in the form &<character name>; see sections
5.3.1 and 5.3.2 of the HTML 4.0 specification.)

This is a change that interests mainly Squeakers in Western Europe.

Jean-Marie tested a first version and found a new bug, later he tested a
second version that is seemingly ok. With his test he helped me to
understand where the real problem was burried. Thanks a lot!

Greetings, Boris


-------------- next part --------------
A non-text attachment was scrubbed...
Name: ISO8859.6.cs
Type: application/octet-stream
Size: 3421 bytes
Desc: not available
Url : http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20021116/2c023663/ISO8859.6.obj